Re: [Teas] AD review: draft-ietf-teas-network-assigned-upstream-label

Deborah, Hi!

We posted a new revision (-09) to address the comments below. Please go
through the diffs (
https://tools.ietf.org/rfcdiff?url2=draft-ietf-teas-network-assigned-upstream-label-09.txt)
and let us know if the changes adequately address the comments.

Please see inline for more responses (prefixed VPB)

Regards,
-Pavan (on behalf of the authors)

On Tue, Aug 22, 2017 at 6:27 PM, BRUNGARD, DEBORAH A <db3546@att.com> wrote:

> Hi,
>
> I've reviewed this draft and after discussing with your Chairs a bit on
> it, I have the following comments:
>
> - I think you need to create an IANA registry for this label if you want
> to ensure it does not get used in an implementation. For example, MPLS has
> registries for "special purpose label values". While this is only one
> value, I suggest a similar top-level registry, with an entry for this draft.
>

[VPB] Added an IANA request for a new subregistry (Please see Section 6).

> - the draft refers to an implementation survey as justifying there are no
> backward compatibility concerns with the choice of this label. But there
> was no implementation survey. I think today's RFC3473 procedures will have
> an upstream node rejecting this label if returned to it and no data traffic
> will be transmitted. So this should prevent any harm. I recommend being
> more concise in the backward compatibility section on the processing.
>

[VPB] Removed the "ambiguous" text in the "Backwards Compatibility" section
(Please see Sections 2.1 [second paragraph] and 2.2).

> - Currently the draft updates RFC3473. I think it also updates RFC3471 and
> RFC6205 (especially as the example in the draft is WSON and this draft
> requires a modification of RFC6205 procedures). And you need to clearly
> identify what it is that you are updating in this document vs. the previous
> documents.
>

[VPB] Updated the "Updates" attribute and added relevant supporting text
(Please see Abstract and Section 1 [second paragraph]).

>
> - The draft is a bit too short:-) You need to have proper sections on
> Procedures and Label Format. The current Processing section is not
> sufficient. There's more in the use case, than in the Processing section.
>

[VPB] Added a figure for the "Label format". Added some text clarifying the
scope of the procedures -- it now explicitly states that the scope is
limited to the exchange and processing of messages between an upstream node
and its immediate downstream node. Added more details to the
error-processing procedures. The draft may still appear to a bit short on
text -- but it is just the nature of the change being proposed (the only
thing that is being proposed here is a new value for the existing
UPSTREAM_LABEL object).

> - Considering the changes for the above, I recommend for your Chairs to
> have a short (1 week) WG Last Call on the updated draft. If can do SOON
> (Adrian's definition:-)), I'll leave it as "AD Evaluation", if drags out,
> I'll send it back to the WG.
>
> Thanks,
> Deborah
>
>
> _______________________________________________
> Teas mailing list
> Teas@ietf.org
> https://www.ietf.org/mailman/listinfo/teas
>